SpeechT5 STT Wav2Vec2
This model is a fine-tuned version of facebook/wav2vec2-base-960h on the Lj-Speech dataset. It achieves the following results on the evaluation set:
- Loss: 644.5502
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 653.9298 | 0.3610 | 50 | 646.8141 |
| 840.6591 | 0.7220 | 100 | 646.6405 |
| 1030.5028 | 1.0830 | 150 | 644.3955 |
| 816.1288 | 1.4440 | 200 | 653.6038 |
| 651.3673 | 1.8051 | 250 | 647.9319 |
| 786.9055 | 2.1661 | 300 | 643.6482 |
| 655.5121 | 2.5271 | 350 | 647.9398 |
| 664.6528 | 2.8881 | 400 | 646.9968 |
| 653.3564 | 3.2491 | 450 | 653.9541 |
| 664.1251 | 3.6101 | 500 | 643.4816 |
| 674.7263 | 3.9711 | 550 | 644.8188 |
| 659.9671 | 4.3321 | 600 | 650.9330 |
| 861.3966 | 4.6931 | 650 | 644.5502 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 15